Salama,
un tout petit groupe ,intéréssé par culture malgache,,
Chacun pouvant au rythme de ses foucades, y déposer son grain de sel,,
échange d'articles divers,
Si vous cherchez un film, un article, un livre sur lequel vous n'arrivez pas à mettre le grapin sur Mada, faites nous signe...sainagasydadabe@gmail.com
An analysis of historical storm data reveals
that the average latitude at which tropical cyclones attain their
maximum intensity has undergone a pronounced shift towards the poles
over the past three decades. See Letter p.349
Considerable attention has been devoted to the regional and
global effects of climate variability and climate change on the
behaviour of tropical cyclones over the past decade or so. Catastrophic
events such as Hurricane Katrina (2005), Cyclone Nargis (2008),
Hurricane Sandy (2012) and Typhoon Haiyan (2013) have led scientists and
non-scientists alike to ask how climate change is affecting the
intensity, frequency and location of tropical cyclones around the globe.
There is a general consensus among experts that anthropogenic warming
will lead to fewer, but more intense, tropical cyclones1.
However, little attention has been paid to understanding long-term
shifts in the geographical location of these cyclones, particularly when
at their peak intensities (Fig. 1).
Figure 1: Global distribution of tropical cyclones at their peak intensities.
The background image is from NASA's
Visible Earth catalogue, and the tropical-cyclone data come from the
National Climatic Data Center's IBTrACS archive10, 11
for the period 1982–2012. Only the locations of storms that achieved an
intensity of at least a category 1 hurricane (that is, a wind speed of
at least 119 kilometres per hour) are shown. The locations represent a
subset of the 'best-track' data used by Kossin and colleagues2 to construct global and regional trends in the mean latitude at which storms reached their maximum intensities.
Background image: NASA Goddard Space Flight Center
On page 349 of this issue, Kossin and co-authors2
shed light on this aspect by examining trends in the latitude at which
the maximum intensities of storms occur, a metric referred to in their
study as the lifetime-maximum intensity (LMI). Their findings reveal a
pronounced migration of the annual-mean LMI towards the poles over the
past 30 years, at a rate of about 1° of latitude per decade, although
this metric varies considerably on regional scales. If this poleward
migration of tropical-cyclone LMI continues, it will probably have major
impacts, including increased threats to coastal communities that have
historically not been susceptible to hazards posed by tropical cyclones.
The
observed poleward trends in the annual-mean LMI are consistent with,
and within the range of, the observed expansion of the tropics since
about 1979 (refs 3, 4).
Several climate-related features have been used to diagnose this
expansion, which is thought to be due to increased concentrations of
anthropogenic greenhouse gases. These features include ozone depletion
in the stratosphere, which lies just above the lowest portion of the
atmosphere (the troposphere); the height of the boundary between the
stratosphere and the troposphere (the tropopause); and the width of the
Hadley circulation, the main meridional overturning circulation in the
troposphere, which is characterized by rising air and thunderstorms near
the Equator and dry, sinking air at around 30° north and 30° south,
where many of the world's deserts are found.
Kossin et al.
suggest that two factors known to modulate tropical-cyclone development
and intensity may have contributed to the observed poleward migration of
annual-mean LMI: deep-layer vertical wind shear, that is, the absolute
difference between wind speeds in the upper and lower troposphere; and
potential intensity, a thermodynamically based theoretical upper limit
of tropical-cyclone intensity that depends on local sea surface
temperature and atmospheric temperature and humidity. Many storms never
achieve their potential intensity because of competing influences, such
as strong vertical wind shear and intrusions of dry air. However, in
principle, increased potential intensity and decreased vertical wind
shear should promote more-intense storms, all other factors being equal.
That such trends moving away from the Equator have been observed over
the past 30 years (see Fig. 2 of the paper2) therefore seems at least qualitatively consistent with the observed poleward migration of annual-mean LMI.
Despite
the large and statistically significant global trends in the
annual-mean latitude of LMI, substantial region-to-region and
year-to-year variability is evident. For instance, the North Atlantic
region, which has received considerable media attention owing to events
such as hurricanes Katrina and Sandy, shows almost no poleward trend on
the basis of historical 'best-track' data over the past 30 years.
Moreover, when the authors used a state-of-the-art data set of
tropical-cyclone intensity (ADT-HURSAT; ref. 5), an opposite, equatorward, trend is found for the North Atlantic (see Table 1 of the paper2).
Such regional differences in trends are probably due to climate modes
that extend in time beyond the period for which accurate satellite-based
data are available.
This is one of the limitations of trend
studies based on satellite-derived estimates of tropical-cyclone
intensity. Although the post-1970s geostationary satellite era is
considered to be the most accurate part of the historical
tropical-cyclone record, the relatively short observation period hampers
the detection of trends influenced by modes of climate variability
whose periodicity spans decades or longer, such as the Pacific Decadal
Oscillation6.
Any such variability implies that regions in which the poleward
migration of annual-mean LMI has been more pronounced over the past 30
years might experience less-pronounced trends in the coming decades, and
vice versa. Even on a global scale, a trend of 1° of latitude per
decade of tropical expansion (that is, a 10° shift per century, assuming
a constant rate of expansion) cannot be sustained without implausible
changes to fundamental physical constraints on the global atmospheric
circulation, such as Earth's rotation rate.
On year-to-year
timescales, variability in tropical-cyclone formation and track is
dominated by the phase of the El Niño–Southern Oscillation (ENSO) — the
episodic warming (El Niño) and cooling (La Niña) of the surface
temperature of the tropical Pacific Ocean. El Niño often promotes an
equatorward migration of tropical-cyclone activity, whereas during La
Niña a poleward displacement is observed7, concomitant with changes in the width and intensity of the Hadley circulation8. It is therefore plausible that any trend in ENSO could project onto trends in tropical-cyclone activity. Kossin et al.
attempt to remove this contribution by accounting for the effect of
ENSO on the linear trend of annual-mean LMI latitude and then examining
the residual data. The poleward migration remains pronounced and
statistically significant, suggesting that ENSO plays only a minor part
in the long-term hemispheric and global trends.
Kossin and
colleagues' findings provide insight into the response of global
tropical-cyclone activity to a changing climate. However, several
questions remain unanswered. For instance, will future changes in wind
patterns cause storms to move towards or away from coastlines9?
What are the key mechanisms driving the observed tropical expansion,
and how do these tie in with factors known to modulate tropical-cyclone
intensity? Such questions remain the subject of future research.
Hartmann, D. L.et al. in Climate
Change 2013: The Physical Science Basis. Contribution of Working Group I
to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change (eds Stocker, T. F. et al.) Ch. 2, 226–229 (Cambridge Univ. Press, 2013).
Hamish Ramsay is at the ARC Centre of Excellence for Climate
System Science and the School of Mathematical Sciences, Monash
University, Victoria 3800, Australia.
Aucun commentaire:
Enregistrer un commentaire